preliminary work for open-source

README.md

@@ -1,7 +1,7 @@
 ## SGL: Scalable Graph Learning
 
 **SGL** is a Graph Neural Network (GNN) toolkit targeting scalable graph learning, which supports deep graph learning on
-extremely large datasets. SGL allows users to easily implement scalable graph neural networks and evalaute its
+extremely large datasets. SGL allows users to easily implement scalable graph neural networks and evaluate its
 performance on various downstream tasks like node classification, node clustering, and link prediction. Further, SGL
 supports auto neural architecture search functionality based
 on <a href="https://github.com/PKU-DAIR/open-box" target="_blank" rel="nofollow">OpenBox</a>. SGL is designed and
@@ -12,21 +12,29 @@ the <a href="https://cuibinpku.github.io/index.html" target="_blank" rel="nofoll
 
 + **High scalability**: Follow the scalable design paradigm **SGAP**
   in <a href="https://arxiv.org/abs/2203.00638" target="_blank" rel="nofollow">PaSca</a>, SGL scale to graph data with
-  billions of nodes and edegs.
+  billions of nodes and edges.
 + **Auto neural architecture search**: Automatically choose decent neural architectures according to specific tasks, and
   pre-defined objectives (e.g., inference time).
 + **Ease of use**: User-friendly interfaces of implementing existing scalable GNNs and executing various downstream
   tasks.
 
 ## Installation (TODO)
 
-#### Install from pip
+Some datasets in SGL are constructed based
+on <a href="https://github.com/pyg-team/pytorch_geometric" target="_blank" rel="nofollow">PyG</a>. Please follow the
+link below to install PyG first before installing
+SGL: https://pytorch-geometric.readthedocs.io/en/latest/notes/installation.html.
 
-#### Install from source
+### Install from pip
+
+To install SGL from PyPI:
+
+```bash
+pip install sgl-dair
+```
 
 ## Quick Start (TODO)
 
-**TODO**
 A quick start example is given by:
 
 ```python
@@ -39,7 +47,6 @@ model = SGC(prop_steps=3, feat_dim=dataset.num_features, num_classes=dataset.num
 
 device = "cuda:0"
 test_acc = NodeClassification(dataset, model, lr=0.1, weight_decay=5e-5, epochs=200, device=device).test_acc
-print(test_acc)
 ```
 
 **TODO**

test.py → examples/sgc_pubmed.py

@@ -1,10 +1,9 @@
-from dataset import Planetoid
-from models.homo import SGC
-from tasks import NodeClassification
+from sgl.dataset import Planetoid
+from sgl.models.homo import SGC
+from sgl.tasks import NodeClassification
 
 dataset = Planetoid("pubmed", "./", "official")
 model = SGC(prop_steps=3, feat_dim=dataset.num_features, num_classes=dataset.num_classes)
 
 device = "cuda:0"
 test_acc = NodeClassification(dataset, model, lr=0.1, weight_decay=5e-5, epochs=200, device=device).test_acc
-print(test_acc)

test_nas.py → examples/test_nas.py

@@ -1,9 +1,9 @@
 import numpy as np
 import torch
 
-from dataset import Planetoid
-from models.search_models import SearchModel
-from search.auto_search import SearchManager
+from sgl.dataset import Planetoid
+from sgl.models.search_models import SearchModel
+from sgl.search.auto_search import SearchManager
 
 dataset = Planetoid("cora", "./", "official")
 

pyproject.toml

@@ -0,0 +1,6 @@
+[build-system]
+requires = [
+    "setuptools>=42",
+    "wheel"
+]
+build-backend = "setuptools.build_meta"

requirements.txt

@@ -1,5 +1,9 @@
-torch==1.11.0
-scipy==1.5.2
-ogb==1.3.3
-numpy==1.22.3
-torch-geometric
+torch>=1.8
+networkx
+tqdm
+numpy>=1.21
+scipy
+gensim
+scikit_learn
+ogb
+openbox

setup.py

@@ -0,0 +1,26 @@
+import setuptools
+
+with open("README.md", "r", encoding="utf-8") as readme:
+    long_description = readme.read()
+
+with open("requirements.txt", "r", encoding="utf-8") as require:
+    requirements = [r for r in require.read().splitlines() if r != '']
+
+setuptools.setup(
+    name="sgl-dair",
+    version="0.1.0",
+    author="DAIR Lab @PKU",
+    description="Graph Neural Network (GNN) toolkit targeting scalable graph learning",
+    long_description=long_description,
+    long_description_content_type="text/markdown",
+    url="https://github.com/PKU-DAIR/SGL",
+    classifiers=[
+        "Programming Language :: Python :: 3",
+        "License :: OSI Approved :: MIT License",
+        "Operating System :: OS Independent",
+    ],
+    packages=setuptools.find_packages(),
+    python_requires='>=3.6',
+    install_requires=requirements,
+    data_files=["requirements.txt"],
+)

sgl/__init__.py

@@ -0,0 +1 @@
+name = "sgl"

data/base_dataset.py → sgl/data/base_dataset.py

@@ -6,9 +6,9 @@
 import warnings
 from scipy.sparse import csr_matrix
 
-from data.base_data import Node, Edge
-from data.utils import file_exist, to_undirected
-from dataset.choose_edge_type import ChooseMultiSubgraphs
+from sgl.data.base_data import Node, Edge
+from sgl.data.utils import file_exist, to_undirected
+from sgl.dataset.choose_edge_type import ChooseMultiSubgraphs
 
 
 # Base class for node-level tasks

dataset/acm.py → sgl/dataset/acm.py

@@ -4,9 +4,9 @@
 from torch_geometric.datasets import HGBDataset
 from typing import Tuple
 
-from data.base_data import HeteroGraph
-from data.base_dataset import HeteroNodeDataset
-from dataset.utils import pkl_read_file
+from sgl.data.base_data import HeteroGraph
+from sgl.data.base_dataset import HeteroNodeDataset
+from sgl.dataset.utils import pkl_read_file
 
 
 class Acm(HeteroNodeDataset):

dataset/actor.py → sgl/dataset/actor.py

@@ -4,9 +4,9 @@
 import torch
 from torch_sparse import SparseTensor, coalesce
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to
 
 
 class Actor(NodeDataset):

dataset/airports.py → sgl/dataset/airports.py

@@ -3,9 +3,9 @@
 import pickle as pkl
 import torch
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to, random_split_dataset
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to, random_split_dataset
 
 
 class Airports(NodeDataset):

dataset/amazon.py → sgl/dataset/amazon.py

@@ -2,9 +2,9 @@
 import pickle as pkl
 import torch
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to, read_npz, random_split_dataset
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to, read_npz, random_split_dataset
 
 
 class Amazon(NodeDataset):

dataset/amazon_product.py → sgl/dataset/amazon_product.py

@@ -6,9 +6,9 @@
 import scipy.sparse as sp
 import torch
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to
 
 
 class AmazonProduct(NodeDataset):

dataset/coauthor.py → sgl/dataset/coauthor.py

@@ -2,9 +2,9 @@
 import pickle as pkl
 import torch
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to, read_npz, random_split_dataset
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to, read_npz, random_split_dataset
 
 
 class Coauthor(NodeDataset):

dataset/dblp.py → sgl/dataset/dblp.py

@@ -3,10 +3,10 @@
 import torch
 from typing import Tuple
 
-from data.base_data import HeteroGraph
-from data.base_dataset import HeteroNodeDataset
-from dataset.dblp_original import DBLP
-from dataset.utils import pkl_read_file
+from sgl.data.base_data import HeteroGraph
+from sgl.data.base_dataset import HeteroNodeDataset
+from sgl.dataset.dblp_original import DBLP
+from sgl.dataset.utils import pkl_read_file
 
 
 class Dblp(HeteroNodeDataset):

dataset/facebook.py → sgl/dataset/facebook.py

@@ -3,9 +3,9 @@
 import pickle as pkl
 import torch
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to, random_split_dataset
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to, random_split_dataset
 
 
 class Facebook(NodeDataset):

dataset/flickr.py → sgl/dataset/flickr.py

@@ -5,9 +5,9 @@
 import scipy.sparse as sp
 import torch
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to
 
 
 class Flickr(NodeDataset):

dataset/github.py → sgl/dataset/github.py

@@ -3,9 +3,9 @@
 import pickle as pkl
 import torch
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import download_to, pkl_read_file, random_split_dataset
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import download_to, pkl_read_file, random_split_dataset
 
 
 class Github(NodeDataset):

dataset/karateclub.py → sgl/dataset/karateclub.py

@@ -4,9 +4,9 @@
 import pickle as pkl
 import torch
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file
 
 
 class KarateClub(NodeDataset):

dataset/linkx_dataset.py → sgl/dataset/linkx_dataset.py

@@ -5,9 +5,9 @@
 import torch.nn.functional as F
 from scipy.io import loadmat
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to
 
 
 # A variety of non-homophilous graph datasets

dataset/nell.py → sgl/dataset/nell.py

@@ -8,9 +8,9 @@
 from torch_geometric.data import extract_tar
 from torch_geometric.io import read_txt_array
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to
 
 
 class Nell(NodeDataset):

dataset/ogbn.py → sgl/dataset/ogbn.py

@@ -4,9 +4,9 @@
 import torch
 from ogb.nodeproppred import PygNodePropPredDataset
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, to_undirected
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, to_undirected
 
 
 class Ogbn(NodeDataset):

dataset/ogbn_mag.py → sgl/dataset/ogbn_mag.py

@@ -4,9 +4,9 @@
 import torch
 from ogb.nodeproppred import PygNodePropPredDataset
 
-from data.base_data import HeteroGraph
-from data.base_dataset import HeteroNodeDataset
-from dataset.utils import pkl_read_file, to_undirected
+from sgl.data.base_data import HeteroGraph
+from sgl.data.base_dataset import HeteroNodeDataset
+from sgl.dataset.utils import pkl_read_file, to_undirected
 
 
 class OgbnMag(HeteroNodeDataset):

dataset/planetoid.py → sgl/dataset/planetoid.py

@@ -5,9 +5,9 @@
 import scipy.sparse as sp
 import torch
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to
 
 
 class Planetoid(NodeDataset):

dataset/reddit.py → sgl/dataset/reddit.py

@@ -6,9 +6,9 @@
 import torch
 from torch_geometric.data import extract_zip
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to
 
 
 class Reddit(NodeDataset):

dataset/twitch.py → sgl/dataset/twitch.py

@@ -3,9 +3,9 @@
 import pickle as pkl
 import torch
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to, random_split_dataset
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to, random_split_dataset
 
 
 class Twitch(NodeDataset):

dataset/webkb.py → sgl/dataset/webkb.py

@@ -4,9 +4,9 @@
 import torch
 from torch_sparse import coalesce
 
-from data.base_data import Graph
-from data.base_dataset import NodeDataset
-from dataset.utils import pkl_read_file, download_to
+from sgl.data.base_data import Graph
+from sgl.data.base_dataset import NodeDataset
+from sgl.dataset.utils import pkl_read_file, download_to
 
 
 class WebKB(NodeDataset):